Aidan Tinnion
Welcome!
In 2024, Oxford's word of the year was "Brain rot".
(n.) Supposed deterioration of a person’s mental or intellectual state, especially viewed as a result of overconsumption of material (now particularly online content) considered to be trivial or unchallenging. Also: something characterized as likely to lead to such deterioration.
Often times, this spread of brain rot covers vast distances and affects many people. So, when I puzzled on ways to explain the spread of diseases, I had to look no further than the people in my life quoting "demure", "skibidi" and "sigma".
The Maths behind "Brain Rot"
During this article, we will study the artificial “brain rot” disease. This refers to the excessive exposure to low quality social media, entailing the use of slang words like “skibidi” and “aura”, typically used by Gen Alpha. Although, brain rot is not spread by pathogens, this still follows the typical spread of a disease but through the words we use on a day-to-day basis. This spread of disease through communication, and the pool of people at risk is increased by social media. Hence, although brain rot is not a legitimate disease, it can still be modelled as one.
Brain rot can be represented as three pools of people: the susceptible, infected and recovered groups, known as the SIR model. The susceptible group are the Gen Alpha kids and some Gen Z who are more inclined to use brain rot terminology. Even some older groups such as millennials may be susceptible to the disease as they’ll try to use it to fit in with younger generations. The infected group refers to those who have contracted the disease and actively use the slang. The final group are those who have recovered from the brain rot disease, for instance the people who have grown out of this phase or have realised it is cringe-worthy.
To aid us in mapping out this terrible disease, we apply the use of mathematical code. This allows us to run programs that can predict the growth of the disease, whilst simultaneously being able to tweak detail. For example, how infectious this disease is, how many people could be infected and the rates of recovery and infection from this disease. Another important choice is the type of model that we would like to use. The options are either a deterministic or a stochastic model. But what is the difference between these?
A deterministic model is a straightforward model that will act the same every time you run the program and hence has a predetermined outcome. However, a stochastic model differs as it is a model based in probability. This means that the outcome is left to chance based on the variables we have input into the program. Therefore, if we take a deterministic model, it will know exactly who will and won’t be infected before running the program. However, in a stochastic model, we use a uniform distribution. A uniform distribution means that every number has the same probability and using this distribution we can randomly generate a number between 0 and 1. Using this value, we compute a calculation with the variables mentioned previously to determine whether someone will be infected, will recover or nothing will happen at that step. We then repeat this step until we reach the end of the program.
For now, let’s focus on only one person. Assume in our first test this person gets infected with brain rot at a later stage of the epidemic and then recovers from the disease after becoming self-aware. In a second test, this same person never becomes infected with the disease because they knew it was cringe from the beginning. Finally, if we take a third test, they become infected right at the beginning of the outbreak and never recover from the disease – I’m sure you know people who are like this.
If there is such a large variety of outcomes based off only one person, increasing the sample size to the whole population would mean even greater outcomes of varying levels. Consequently, this could mean that sometimes the disease will be predicted to instantly die out and in the next simulation it may blow up to infect everyone.
Then how do we gain anything useful from this? To benefit from the stochastic model, we run the model numerous times to find an average of the distributions. This new average distribution is now a probability distribution and can be used to predict what will happen. In the real world, there is no certainty with how a disease will develop. Although a stochastic model is more time-consuming than a deterministic model, the stochastic model is typically better to model a disease because it can include the unpredictable spread that real diseases have.
Now, if we apply the stochastic model to the usage of the word “skibidi”, like most diseases, initially the number of infected people is low, and the growth is minimal. However, as time passes it gains more traction, and the rate of infection increases. This is around the time of when people have started hearing the word but not yet using it themselves. Next, the number of people infected catapults as the growth increases massively. “Skibidi” is now being used by many people and the number of people infected with brain rot begins to reach its peak. At this point, the word has become embarrassing to use and as the group of infected people realise this negative connotation they stop using the word. This creates a shift of infected people moving to the recovered group in the model. Hence, the number of infected begin to plumet as there are less susceptible people to maintain the growth of the disease. Therefore, by using the stochastic model, I predict that “skibidi” will have died out by 2026 and it will be nothing more than a cringe-worthy memory.